Tag: agent benchmarking
Browse our exclusive articles!
FinTrace: Evaluating LLM Tool Use for Financial Tasks
Discover FinTrace, a benchmark for assessing LLM tool-calling in long-horizon financial tasks with detailed trajectory-level evaluation metrics.
Spatial Competence Benchmark for AI Model Evaluation
Discover SCBench, a novel benchmark assessing AI spatial reasoning with executable tasks and robust verification for advanced model evaluation.
MobiFlow: Real-World Mobile Agent Benchmarking Framework
Discover MobiFlow, a novel framework for benchmarking mobile agents using trajectory fusion to enhance real-world task evaluation across apps.
Turing Test on Screen: Benchmarking Mobile GUI Agent Humanization
Explore the Turing Test on Screen benchmark for enhancing mobile GUI agents' human-like behavior to improve detection and interaction in digital environmen...